Anomaly Detection of Time Series with Smoothness-Inducing Sequential Variational Auto-Encoder

TL;DR

This paper introduces SISVAE, a novel deep generative model combining RNNs and VAEs with a smoothness prior, for robust anomaly detection in multi-dimensional time series, outperforming existing methods on synthetic and real data.

Contribution

It proposes a smoothness-inducing prior within a sequential VAE framework to enhance anomaly detection robustness in time series.

Findings

Effective anomaly detection on synthetic datasets.

Improved performance on public real-world benchmarks.

Robust density estimation without assuming constant noise.

Abstract

Deep generative models have demonstrated their effectiveness in learning latent representation and modeling complex dependencies of time series. In this paper, we present a Smoothness-Inducing Sequential Variational Auto-Encoder (SISVAE) model for robust estimation and anomaly detection of multi-dimensional time series. Our model is based on Variational Auto-Encoder (VAE), and its backbone is fulfilled by a Recurrent Neural Network to capture latent temporal structures of time series for both generative model and inference model. Specifically, our model parameterizes mean and variance for each time-stamp with flexible neural networks, resulting in a non-stationary model that can work without the assumption of constant noise as commonly made by existing Markov models. However, such a flexibility may cause the model fragile to anomalies. To achieve robust density estimation which can also benefit detection tasks, we propose a smoothness-inducing prior over possible estimations. The proposed prior works as a regularizer that places penalty at non-smooth reconstructions. Our model is learned efficiently with a novel stochastic gradient variational Bayes estimator. In particular, we study two decision criteria for anomaly detection: reconstruction probability and reconstruction error. We show the effectiveness of our model on both synthetic datasets and public real-world benchmarks.